Revanth Bodepudi and Prerit Oberai are betting that artificial intelligence can do what decades of automation couldn't: turn the messy, slow-moving world of custom part manufacturing into something approaching an on-demand service.
Their startup, Prototyping.io, emerged from Y Combinator's Spring 2026 batch with a pitch that sounds straightforward—upload a CAD file, get a finished mechanical part in 24 hours—and an ambition that verges on audacious. They want to automate not just the quoting and feedback that bog down hardware development, but eventually the production itself. AI-guided robots building parts with minimal human intervention. That sort of thing.
It's the kind of vision that makes sense on a whiteboard. The mechanical parts industry processes over a trillion dollars in orders annually, yet—according to the founders—still operates much like it did before smartphones existed. Engineers submit designs, wait days for quotes, endure back-and-forth emails about manufacturability issues, then cross their fingers that the parts show up on time and within tolerance. Prototyping.io opened early access in early May, promising to collapse that timeline into a single day.
Whether they can actually pull it off is another question entirely.
The Platform, In Theory
The company's system is designed to ingest CAD files and route them through what the founders describe as an AI-driven production pipeline. Automated feature extraction identifies what needs to be machined or formed. Manufacturability analysis flags potential problems before metal gets cut. Production planning determines the optimal process and factory routing. Then—and here's where it gets interesting—manufacturing execution, which in most competitor platforms still means "send it to a human shop manager," is supposed to happen with increasing autonomy.
On paper, the capabilities are comprehensive. CNC machining, sheet metal fabrication, 3D printing, injection molding, extrusion, die casting. Tolerances down to ±0.0002 inches. An expansive materials catalog: aluminum alloys, tool steels, titanium, high-performance polymers like PEEK and PTFE. Surface finishes ranging from basic bead blasting to anodizing and heat treatment. No minimum orders required.
It's an impressive spec sheet. But Prototyping.io is currently operating on a waitlist, and the company hasn't named a single customer publicly as of mid-May 2026. The website claims they're serving "early-stage startups to multi-billion dollar enterprises," which—absent case studies or testimonials—is the kind of assertion that raises more questions than it answers. Serving how many customers? Delivering what volumes? Meeting specifications how consistently?
These details matter in an industry where a missed tolerance can scrap an entire production run and torpedo a product launch timeline.
Who's Behind It
Oberai's background offers some credibility, if not direct manufacturing expertise. He spent time at Microsoft working on Excel and Office Security after a brief stint in a University of Illinois engineering PhD program (he left after 18 months). More recently, he was a founding engineer at an early-stage patient navigation startup before pivoting to manufacturing infrastructure.
Bodepudi's professional history is less publicly documented, though his recent LinkedIn activity suggests an engineer fixated on eliminating the manual chokepoints that plague hardware development. Both founders graduated from Y Combinator's accelerator program this year, working under partner Nicolas Dessaigne at the Sunnyvale office.
Standard YC funding appears to be their main backing so far—Dealroom lists a $125,000 infusion from the accelerator in March 2026, though early-stage funding figures are often incomplete or understated. No additional venture rounds have been announced as of mid-May 2026.
A Crowded, Suddenly AI-Obsessed Field
Prototyping.io isn't pitching into a vacuum. The industry's incumbents have spent the past few months racing to bolt AI capabilities onto their platforms, sometimes with more marketing energy than technical substance.
Protolabs, the publicly traded giant, launched its ProDesk platform earlier this year, promising AI-enhanced quoting and design-for-manufacturing feedback across multiple processes. Xometry has been equally vocal about machine learning deployment for pricing algorithms, manufacturability checks, and supplier routing. RapidDirect announced an AI-powered DFM tool in late March. Even newer players like Phasio and Buildables are marketing "AI copilots" for manufacturing workflows.
Most of these efforts focus on the front end—smarter quoting, faster feedback loops, better design optimization before parts hit the shop floor. Prototyping.io's founders are signaling something more ambitious: what they call "physical AI models for manufacturing" and eventual "autonomous execution through industrial robotics." The implication is that they're not just digitizing the paperwork around manufacturing. They want to automate the actual making of things.
That's where vision meets reality in ways that tend to be unforgiving.
The Credibility Gap
There's no pricing listed on the website. No published service-level agreements. No detailed case studies showing how the platform handled a complex part with challenging tolerances, or what happened when something went wrong. The company's public footprint consists of a website, a Y Combinator Launches post, and periodic founder updates on LinkedIn. No trade press coverage has emerged. No industry veterans vouching for the technology.
For a company targeting hardware teams in robotics, AI infrastructure, and energy—sectors where Prototyping.io says it's focusing—this kind of opacity creates friction. These industries don't run on promises. They run on verified track records, ISO certifications, and the kind of boring operational consistency that keeps supply chains humming.
The founders updated their privacy policy and terms of service in late February, which suggests they formalized operations not long before launching early access. In startup terms, they're essentially brand new. Two people. A waitlist. A big idea about reimagining a trillion-dollar industry.
And perhaps an underestimation of how slowly trust accumulates in manufacturing.
What Comes Next
The value proposition, at least in theory, is clear enough. Hardware teams live and die by iteration speed. If Prototyping.io can genuinely deliver production-quality parts in 24 hours with consistent tolerances and reasonable pricing, they'd find customers. Plenty of them.
But "if" is doing heavy lifting there. The mechanical parts business is littered with startups that underestimated the complexity of coordinating supply chains, maintaining quality across different processes, and managing the thousand small details that separate a good part from an unusable one. Speed matters. But speed without reliability is just expensive chaos.
The one-day promise is bold, maybe bold enough to stand out in a field suddenly crowded with AI-augmented competitors. Now Bodepudi and Oberai face the considerably harder task of proving they can deliver it—not once, not ten times, but repeatedly, at scale, with the kind of boring consistency that turns early adopters into long-term customers.
Machine shop reality has a way of humbling ambitious promises. We'll see if theirs holds up.
